The hypothesis that elaboration of the phobic stimulus potentiates exposure in vivo treatment was put to the test in a clinical experiment. Forty-one female spider phobics, who applied for treatment, were randomly allocated to one of two conditions. In both conditions a 2.5 hour therapist-directed exposure treatment was given. In the elaboration condition Ss were required to attend to and describe the objective features of the spider constantly. In the non-elaboration condition the therapist tried to prevent this. A manipulation check suggested that Ss in the elaboration condition had indeed elaborated more extensively on the objective features of the spiders during exposure than Ss in the non-elaboration condition. However, Ss in the elaboration condition practised with somewhat fewer spiders than Ss in the non-elaboration condition, probably because of the time the elaboration took. Contrary to the hypothesis, elaboration did not potentiate either short-term or long-term effects of the exposure treatment. It seems superfluous to let patients elaborate on the phobic stimulus during exposure in vivo treatment: the processing of phobic stimulus information which is needed for an effective treatment seems to occur spontaneously when the patient engages in exposure in vivo exercises.